FIG. 1 shows a shadow mask 2 and a viewing faceplate 18 of a conventional CRT having a screen assembly 22 thereon. The shadow mask 2 includes a plurality of rectangular openings 4, only one of which is shown. The screen assembly 22 includes a light-absorbing matrix 23 with rectangular openings in which blue-, green-, and red-emitting phosphor lines, B, G, and R, respectively, are disposed. Three color-emitting phosphors and the matrix lines, or guardbands, therebetween comprise a triad having a width or screen pitch, p, of about 0.84 mm (33 mils). The guardbands are designated hereinafter as RB, for the guardbands between the red- and blue-emitting phosphor lines; RG, for the guardbands between the red- and green-emitting phosphor lines; and BG, for the guardbands between the blue- and green-emitting phosphor lines. For the conventional shadow mask 2, the mask openings 4 have a width, a, not greater than one third the width, p, of the triad. In a CRT having a diagonal dimension of 51 cm (20 inches), the width, a, of the shadow mask openings 4 are on the order of about 0.23 mm (9 mils) and the resultant openings formed in the matrix have a width, b, of about 0.18 mm (7 mils). The guardbands of the matrix 23, between the adjacent phosphor lines, have a width, c, of about 0.1 mm (4 mils). The matrix 23, preferably, is formed on the viewing faceplate 18 by the process described in U.S. Pat. No. 3,558,310, issued to Mayaud on Jan. 26, 1971. Briefly, a film of a suitable photoresist, whose solubility is altered by light, is provided on the viewing faceplate. The photoresist film is exposed, through the openings 4 in the shadow mask 2, to ultraviolet light from a conventional three-in-one lighthouse, not shown. After each exposure, the light is moved to a different position, within the lighthouse, to duplicate the incident angles of the electron beams from the electron gun of the CRT. Typically, the three electron beam positions, designated 6, 7 and 8, are spaced a distance, X.sub.0, about 5.38 mm (212 mils) apart, as shown in FIG. 2. Three exposures are required, from the three different lamp positions, to complete the matrix exposure process. Then, the regions of the film with greater solubility are removed by flushing the exposed film with water, thereby uncovering bare areas of the faceplate panel. Next, the interior surface of the faceplate panel is overcoated with a black matrix slurry, of the type known in the art, which, when dried, is adherent to the uncovered areas of the faceplate panel. Finally, the matrix material overlying the retained film regions, as well as the retained film regions, are removed, leaving the matrix layer on the previously uncovered areas of the faceplate panel. Again with reference to FIG. 1, the difference between the width, a, of the shadow mask openings and the width, b, of the matrix openings is referred to as "print down." Thus, in the conventional shadow mask-type CRT of FIG. 1, having mask openings with a width of 0.23 mm and matrix openings with a width of 0.18 mm, the typical "print down" is about 0.05 mm (2 mils). A drawback of the shadow mask-type CRT is that, at the center of the screen, the shadow mask intercepts all but about 18-22% of the electron beam current; that is, the shadow mask is said to have a transmission of only about 18-22%. Thus, the area of the openings 4 in the shadow mask 2 is about 18-22% of the area of the mask. Because there are no focusing fields associated with the shadow mask 2, a corresponding portion of the screen assembly 22 is excited by the electron beams.
In order to increase the transmission of the color selection electrode without increasing the size of the excited portions of the screen, a post-deflection focusing color selection structure is required. The focusing characteristics of such a structure permit larger aperture openings to be utilized to obtain greater electron beam transmission than can be obtained with the conventional shadow mask. One such structure, a uniaxial tension focus mask, is described in U.S. Pat. No. 5,646,478 issued to R. W. Nosker et al. on Jul. 8, 1997. A drawback of using a post deflection color selection electrode, such as a tension focus mask, is that conventional methods for forming the matrix cannot be utilized, because the prior methods provide only about a 0.05 mm (2 mil) "print down." For the tension focus mask of U.S. Pat. No. 5,646,478, the triad period, p, of the screen assembly is the same as for a CRT with a conventional shadow mask, so the matrix openings are about 0.18 mm wide. However, as described hereinafter, for a tension focus mask-type CRT, a "print down" of about 0.37 mm (14.5 mils) is required. Such a high degree of "print down" cannot be achieved with the conventional matrix process described above. Additionally, for a tension focus mask-type CRT, any matrix opening patterns formed using a conventional three-in-one lighthouse process, such as that taught by Mayaud, referenced above, will result in misregister of the electron beams which impinge upon the blue- and red-emitting phosphors with "Q"-space errors. The dimension "Q" is the distance between the color selection electrode and the inner surface of the faceplate. "Q"-space errors of the order of +/-5%, that is variations in the focus mask-to-screen spacing caused by deviations of the faceplate thickness or curvature from the bogie dimensions, are typical. Accordingly, a new method of making a matrix with the capability for very large "print down" with no electron beam misregister is required.